It’s no secret that the American education system is saturated with standardized tests.
As of 2016, the average student in America takes a staggering 112 mandatory standardized tests before graduating high school. This averages out to be eight a year according to one 2-year study by the Council of Great City Schools1.
The tests are “standardized” because all students answer the same questions under comparable conditions and their responses are scored using measured criteria, whether the questions are multiple choice or open-ended2.
However, there are numerous reasons to believe that high stakes standardized tests are actually quite damaging to education and have received forceful criticism over the past dozen years as a result. Examples include their propensity to drive out teachers, encouraging teaching “to the test” as well as increasing grade retention and school dropout rates, all of which question the imposition of high quantities of standardized tests throughout a student’s school career2.
Furthermore, the disadvantages that standardized tests pose for the many students who take them are substantial. Although these tests were conceptualized to ensure fairness and equity for all, the reality is much more grim. In addition to the problematic application for students with diagnosed (and undiagnosed) learning disabilities and to non-native English speakers, these tests are unfair to countless others due to a host of social, cultural, economic and even biological reasons. One example is the marked disadvantage for students from underprivileged groups, for which there is a proven departure between test scores and actual academic potential3.
In other words, standardized tests can undermine the very abilities they seek to assess.
The goal of this article is to look “under the hood” into some of the research investigating what occurs in the brains and minds of test takers, with an emphasis on studies that employ neurophysiological (e.g. EEG) and functional neuroimaging (e.g. fMRI) methods. These methods allow researchers to look inside the working brain to explore the neural underpinnings of the psychological processes that so many students experience during test taking. The hope is that, by better understanding these mechanisms, we can craft better environments for teaching and evaluating our students.
This article will be limited to discussing neuroscience research on three interrelated psychological processes that occur during standardized test taking, namely i) testing anxiety, ii) “choking” under pressure and iii) stereotype threat. However, it should be noted that this is by no means a comprehensive list and should be viewed as a small glimpse into what happens in the brain during high-pressure standardized testing.
The consequences of falling into socially constructed stereotypes
Over the past twenty years, stereotype threat has become one of the most investigated topics in social psychology. It invokes the age-old nature vs nurture debate, as well as questions related to group differences observed in cognitive performance.
Broadly speaking, stereotype threat can be defined as a situation in which individuals believe they are at risk for confirming negative beliefs about their social group. It rose to prominence via a pioneering 1995 study by Steele and Aronson4, which found that African American student performance varied depending on the immediate messaging from the environment. Aware of the negative stereotypes regarding their intellectual abilities and thus fearing confirming the stereotype, students exhibited suboptimal performance when asked to display such abilities (in the case of Steele and Aronson, a difficult verbal test) after being reminded of their stereotyped group membership. African American students performed better in a control condition, in which they were not primed with the threat. As a result, the concept is considered highly salient within education.
These findings have been replicated by many others over the years and are generalizable to other stereotyped groups, including (but not limited to) women’s mathematical abilities5 and memory recall abilities in elderly populations6. Importantly, enthusiasm for this work is bolstered by the notion that these findings help explain the discrepancies observed in academic and cognitive performance among at-risk groups7, undermining the assumption that the differences are a result of innate (in)ability.
In an attempt to review the reasons behind impaired performance, one review article8 revealed patterns of activation demonstrating that individuals “under threat” have heightened activity in parts of the brain devoted to regulating emotions. Simultaneously, they showed decreased activation in areas typically associated with executive function (such as working memory and attention).
In another study9 evaluating a negative stereotype regarding women’s spatial reasoning skills, researchers found that in addition to activation in areas of the brain associated with spatial rotation, participants in the threat condition also had increased activity in regions of the brain associated with emotion regulation and social knowledge (e.g. the rostral-ventral anterior cingulate cortex and right orbital gyrus) synonymous with feelings of anxiety regarding one’s social group.
These results are consistent with a study from 2014 using EEG10 – a physiological measure favored for its ability to secure very accurate readings of when electrical activity occurs in the brain – which investigated the effect of stereotype threat on women’s performance on a series of math problems. Here, the authors reported that the threat condition negatively affects the anterior cingulate cortex and the dorsolateral prefrontal cortex, which was not found in women placed in the condition in which the stereotype threat was not activated. In other words, the bias caused by the stereotype threat had damaging consequences for attention, memory and general information processing – systems highly likely to play a key role in undermining performance in the math task.
Collectively, what these studies illustrate is that participants under stereotype threat are hypothesized to do worse due to the allocation of important cognitive resources to things like emotion regulation and social knowledge as opposed to concentrating on the task at hand.
Although these studies test abilities within an isolated experimental environment, educational neuroscientists studying the effects of high-stakes testing argue that even when not primed, at-risk groups “underperform relative to their ability merely because they are aware of a negative stereotype about how they should perform”11. An example of these worries can be seen in the female math test-taker who is apprehensive given the stereotype that “guys are better than girls at math”11.
“Choking” under pressure & testing anxiety
Another dire consequence of high-stakes testing is choking under pressure, a phenomenon associated with poor performance on a cognitive task and one easily relatable to standardized test taking. This is the tendency for some individuals to have their performance hindered by the stress of a given situation. In the case of standardized testing, many controlled laboratory experiments have found evidence of this on a wide range of tasks, most notably math problem solving, category learning, and tests of fluid intelligence.12,13,14
But what are the mechanisms behind choking under pressure? Researchers have detailed three (not mutually exclusive) explanations for why testers choke under pressure; i) the “distraction account” whereby the pressure of doing well distracts the individual from the task(s) at hand, ii) the “over-monitoring account”, when task performance is worsened due to a hyper vigilance in attending to every required step and finally iii) the “over-arousal account” leaving the person in a heightened, and often stressed, emotional state due to the lingering fear of large losses.15, 16
The Distraction Model vs. The Over-Monitoring Model
The “distraction” model can be understood in terms of a distraction caused by the pressure to do well on a demanding task (e.g. math problem solving), making it difficult for a person’s working memory to perform optimally. By contrast the “over-monitoring” model argues the opposite of the “distraction” model: instead of being distracted by the pressure to do the task at hand, this theory argues that the extra attention, control and motor operation given to the task actually hinders performance16. In other words, focusing too much on specific details of a task can take away from the cognitive “horsepower” (e.g. executive function) needed to complete the task successfully.
Although these theories diverge, they are not necessarily mutually exclusive; both can occur simultaneously depending on the nuances and demands of the task as well as the source(s) of motivation provided. Indeed, both theories are supported by fMRI evidence revealing a statistically meaningful relationship between choking and compromised executive control in prefrontal areas of the brain.
Strikingly, further analyses in this area has suggested that individuals with higher working memory capacities are actually even more susceptible to this kind of stress than people with lower working memory capacities, given their reliance on this strength to perform well.17
The Over Arousal Model
The third model concerns the neural mechanisms behind reward and motivation. One theory favored by behavioral economists suggests that degraded performance is a result of the flooding of emotion due to high pressure or incentive. You can think of it along the lines of being stressed out by thinking about how big the stakes are. To make matters worse for standardized test takers, many studies have shown that performance is only diminished when the task at hand is one that is complex or not well-learned, such as the ones required during test taking18, 19. By contrast, this phenomenon is not observed during very simple conditioning tasks such as those that require considerable physical effort (e.g. weight lifting), whereby the increased incentive or social pressure can actually enhance performance.20
From a neural perspective, one study found significant deactivation in areas of the brain that are crucial for memory recall as well as hormone and emotion regulation (e.g. the hippocampus, hypothalamus, medial orbital cortex and ACC) in participants subjected to the stress condition. These participants also displayed increased levels of cortisol, which is linked to heightened stress or emotion21. Specifically, the extent of deactivation in the parts of the brain implicated in memory was strongly related to the release of cortisol in response to a stress task.
A similar yet distinct notion common amongst standardized test takers is that of test anxiety, which can come into play during any of the situations discussed previously. Generally speaking, test anxiety refers to an unpleasant emotional and physiological reaction that can include feelings of worry, fear of failure or dread before or during a test. The two key features of test anxiety are thought to be i) emotionality (i.e. feeling anxious) and ii) worry. This is corroborated by neuroimaging evidence showing test anxiety to involve enhanced attention devoted to the threat and poor ability to control the feelings of threat.
Some theorists have conceptualized test anxiety as a dispositional factor, whereby “test trait anxiety” can be considered a trait of a person’s personality that make them predisposed to experience greater-than-average levels of stress or anxiety during a testing situation22. Others have defined it as a genuine phobia, based on the elevated physiological responses linked to heightened emotion and stress23.
Regardless of its characterization, test anxiety (with math anxiety specifically being the most common) is highly prevalent, with one source claiming it affects an estimated 25% of 4-year college students and up to 80% of community college students in the U.S.24. One such sources comes from outside the U.S., where researchers from the University of Grenada found 6 in 10 Spanish university students to suffer from math anxiety, with greater incidence among women compared to men25.
Crucially, a fascinating 2011 study showed math anxiety doesn’t just occur during math problem solving, but also in anticipation of doing math. Using fMRI, the researchers were able to detect differences in brain activation between the anticipation stage and actually doing the math task26. What they found was that high-anxiety individuals showed increased fronto-parietal network, which is linked with control of negative emotions. In a follow-up study, they revealed that this anticipation stage is also associated with pain networks in the brain (e.g. mid-cingulate cortex and insula).
This suggests that for some people, merely being faced with the prospect of doing math can be psychologically painful27.
To muddy the waters, reports have pointed to the similarities between stereotype threat and test anxiety and how they may overlap11. Although further research is needed to pinpoint interactive effects, recent studies have indicated test anxiety and stereotype threat might share certain mechanisms within the brain, such as their mutual negative impact on working memory. Furthermore, it has been found that stereotype threat perpetuates states of worry, with worry being a key element in test anxiety. However their interaction is a complex one not yet fully fleshed out in the literature22.
Despite this evidence, it stands today that the primary objective of standardized tests is to reliability measure a student’s academic potential.
So, now what?
So if we shouldn’t use standardized tests, then what should we use?
It is established that standardized tests can be incredibly debilitating, not only for students with disabilities or those from underprivileged backgrounds, but also for many others who suffer from the high-stakes pressure and/or testing anxiety. Yet individual differences along with the complex interplay of social-economic status, culture, social pressure, motivation and reward are all important determinants of performances success or failure. Based on the research that is currently available, it’s hard to dispute the notion that standardized tests are a far from perfect measure of academic ability and assessment, with many students being unfairly penalized and undermined for things that they cannot control.
Many have proposed plausible solutions to the standardized test epidemic. Examples include i) simply reducing the number of standardized tests given, ii) replacing data from assessments with data collected “passively” over long periods of time iii) increasing the prevalence of game-based assessments, or iv) implementing more social and emotional skill surveys. Further details on these proposals can be found in the 2015 book by Anya Kamenetz.
As the field of educational neuroscience continues to expand, it behooves the test prep industry to take into consideration the latest research on how the brain works under test conditions. In addition, given the relatively small number of students with the background and resources necessary to achieve very high scores on standardized tests, the scientific markers discussed in this article should play an active role in redefining ability and educational assessment, in the hopes that one day, educational decisions will never be based solely on a number.
References & Further Reading
- Strauss, V. (2015, October 24). Confirmed: Standardized testing has taken over our schools. But who’s to blame? The Washington Post. [Article]
- The National Center for Fair & Open Testing. (2007, December 17). The Dangerous Consequences of High-Stakes Standardized Testing. [Article]
- Rooks, N. (2012, October). Why It’s Time to Get Rid of Standardized Tests, [Article]
- Steele C.M., Aronson J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69:797–811. [Paper]
- Spencer, S. J., Steele, C. M., & Quinn, D. M. (1999). Stereotype threat and women’s math performance.Journal of Experimental Social Psychology, 35, 4–28. [Paper]
- Hess, T.M., Auman, C., Colcombe, S.J. Rahhal, T.A. (2003). The impact of stereotype threat on age differences in memory performance. Journals of Gerontology: Psychological Sciences and Social Sciences, 58(B):3–11. [Paper]
- Schmader, T., Johns, M., & Forbes, C. (2008). An integrated process model of stereotype threat effects on performance. Psychological Review, 115, 336–356. [Paper]
- Derks, B., Inzlicht, M., & Kang, S. K. (2008). The neuroscience of stigma and stereotype threat. Group Processes & Intergroup Relations, 11, 163–181. [Paper]
- Wraga, M., Helt, M., Jacobs, E., & Sullivan, K. (2007). Neural basis of stereotype-induced shifts in women’s mental rotation performance. Social Cognitive and Affective Neuroscience, 2, 12–19. [Paper]
- Forbes, C.E., Leitner, J.B. (2014). Stereotype threat engenders neural attentional bias toward negative feedback to undermine performance. Biological psychology, 102, 98–107. [Paper]
- Maloney, E.A., Schaeffer, M.W., & Beilock, S.L. (2013). Mathematics anxiety and stereotype threat: shared mechanisms, negative consequences and promising interventions, Research in Mathematics Education, 15:2, 115-128. [Paper]
- Beilock, S.L., Carr, T.H. (2005). When high-powered people fail: working memory and “choking under pressure” in math. Psychological Science, 16, 101–105. [Paper]
- Markman, A.B., Maddox, W.T., Worthy, D.A. (2006). Choking and excelling under pressure. Psychological Science, 17, 944–948. [Paper]
- Gimmig, D., Huguet, P., Caverni, J.P., Cury, F. (2006). Choking under pressure and working memory capacity: when performance pressure reduces fluid intelligence. Psychonomic Bulletin Review, 13, 1005–1010. [Paper]
- Yu, R. (2015). Choking under pressure: the neuropsychological mechanisms of incentive-induced performance decrements. Frontiers in Behavioral Neuroscience, 9, 19. Doi: http://doi.org/10.3389/fnbeh.2015.00019 [Paper]
- Lee T. G., Grafton S. T. (2015). Out of control: diminished prefrontal activity coincides with impaired motor performance due to choking under pressure. Neuroimage,105, 145–155. 10.1016/j.neuroimage.2014.10.058 [Paper]
- Mattarella-Micke A., Mateo J., Kozak M. N., Foster K., Beilock S. L. (2011). Choke or thrive? The relation between salivary cortisol and math performance depends on individual differences in working memory and math-anxiety. Emotion, 11, doi: 1000–1005. 10.1037/a0023224 [Paper]
- Yerkes RM, Dodson JD (1908). The relation of strength of stimulus to rapidity of habit-formation. Journal of Comparative Neurology and Psychology, doi:18: 459–482.doi:10.1002/cne.920180503 [Paper]
- Eysenck, M. W., & Calvo, M. G. (1992). Anxiety and performance: The processing efficiency theory. Cognition and Emotion, 6, 409 – 434. [Paper]
- Strauss, B. (2002). Social facilitation in motor tasks: A review of research and theory. Psychology of Sport and Exercise, 3, 237 – 256. [Paper]
- Pruessner J. C., Dedovic K., Khalili-Mahani N., Engert V., Pruessner M., Buss C., et al. . (2008). Deactivation of the limbic system during acute psychosocial stress: evidence from positron emission tomography and functional magnetic resonance imaging studies. Biological Psychiatry, 63, 234–240. 10.1016/j.biopsych.2007.04.041 [Paper]
- Temple, T., & Neumann, R. (2014). Stereotype threat, test anxiety, and mathematics performance. Social Psychology of Education, 17, 491-501 [Paper]
- Faust, M.W. (1992). Analysis of physiological reactivity in mathematics anxiety. Unpublished doctoral dissertation, Bowling Green State University, Bowling Green, Ohio
- Adelson, R. (2014). Nervous About Numbers: Brain Patterns Reflect Math Anxiety. Observer: Association for Psychological Science, 27, September. [Article]
- University of Granada. (2009, April 2). Six Out Of 10 University Students Have Math Anxiety, Spanish Study Finds. ScienceDaily. [Article]
- Lyons, I. M., & Beilock, S. L. (2011). Mathematics anxiety: Separating the math from the anxiety. Cerebral Cortex, 22(9): 2102-10. [Paper]
- Lyons, I. M., & Beilock, S. L. (2012). When Math Hurts: Math Anxiety Predicts Pain Network Activation in Anticipation of Doing Math. PLOS ONE.7(10): e48076. doi:10.1371/journal.pone.0048076 [Paper]